Airdrop, Which Harvests Moisture Directly From Desert Air, Wins James Dyson Award

The James Dyson Award winners for 2011 have been announced, and the grand prize winner is a piece of clever biomimicry that sits so perfectly in our wheelhouse that we couldn’t resist the urge to write about it. Edward Linacre of Swinburne University of Technology in Melbourne has tapped the Namib beetle--a desert dwelling species that survives in the most arid conditions on Earth--to create an irrigation system that can pull liquid moisture straight out of dry desert air.

Airdrop, as the system is known, borrows a trick from the Namib beetle, which can live in areas that receive just half an inch of rain per year by harvesting the moisture from the air that condenses on its back during the early morning hours. A hydrophilic skin helps to snare water molecules passing on the breeze, which then accumulate into droplets of consumable liquid water.

Airdrop mimics this idea, though on a larger scale. The self-powering device pumps water into a network of underground pipes, where it cools enough for water to condensate. From there the moisture is delivered to the roots of nearby plants. Linacre’s math shows that about 11.5 milliliters can be harvested from every cubic meter of air, and further development could raise that number even higher.


Such a system could provide regular moisture to plants being grown in the world’s driest regions. And because it is low cost and self-powered, there’s not a lot of investment or maintenance involved in deploying Airdrop. The $14,000 award from Dyson (Linacre’s university also gets an additional $14,000) should help speed that along.

This year’s runners up included a quickly deployable divider for medical settings that lets healthcare professionals make the most of available space and an aide for the blind that uses a special cane and location-based social networking apps to help the visually impaired locate their friends.

New Li-ion Battery Design Boosts Energy Capacity and Charge Rate 10-Fold

For years, battery designers have been looking for the next big thing in energy storage technology that could replace the lithium-ion batteries currently found in everything from laptops to smartphones to cars. It turns out they may have simply needed to rethink the existing li-ion battery. Northwestern University researchers have re-engineered a lithium-ion battery that can hold ten times the charge of current batteries on the market, and can charge ten times faster.

The trick: a redesigned anode that addresses the two main issues holding li-ion batteries back--charge capacity and charge rate. Li-ion batteries work via a chemical reaction in which lithium ions are swapped between two ends of a battery (known as the anode and the cathode). As energy is burned by a device, ions travel from where they are stored in the anode through an electrolyte to the cathode. In the process, electrical charge is passed to the device as the ions make the transition through the electrolyte. When the battery charges, the ions move in the opposite direction, from cathode to anode.

Current anode design is based on graphene sheets--one-atom-thick layers of carbon--that store the lithium ions. But these anodes can only store one lithium atom for every six carbon atoms, a rather low charge density. Designers have experimented with materials like silicon, which can hold four lithium atoms for every silicon atom, but silicon tends to expand and contract significantly during the charge process, causing it to fragment. This naturally reduces the lifetime of the anode.


A graphene-based design also slows the charge rate. Because of the geometry of graphene sheets--very thin but very long--lithium ions have to make a long trip to the edges of the graphene sheets and then push their way inside. This causes a kind of ion bottleneck around the edges of the anode and slows the charge rate significantly.

The NU team sawed through these problems significantly by rethinking the anode and incorporating a hybrid graphene-silicon design that boosts capacity and charge rate at the same time. First, they sandwiched layers of silicon in between the graphene sheets, allowing greater numbers of lithium ions to come to rest there. The silicon still expands and contracts during charging and discharging, but the flexibility of the graphene still holds the anode together. The silicon can fragment but it still stays in place, allowing the anode to hold greater charge.

The team then used chemical oxidation to punch tiny holes in the graphene sheets--just 10 to 20 nanometers across--so the lithium ions can move through the graphene rather than having to go around to the edges of the anode (where the traffic jams were occurring). This shortcut allow lithium ions to pile into the anode quickly during the charge process, giving charge rates a 10-fold shot in the arm.

And that’s just the anode. The researchers next plan to rethink the cathode to further boost efficiency and effectiveness. The better li-ion battery could hit the marketplace in the next three to five years.

New Material Promises Faster Chips, Faster Internet, Faster Everything

We’ve constructed a world out of fiber optic cable and silicon, but Arizona State University researchers think their new material can do better. They have synthesized a new kind of single-crystal nanowire from a compound of erbium--a material generally used to dope fiber optic cables to amplify their signals--and they claim it could increase the speed of the Internet, spawn a new generation of computers, and improve photovoltaic solar cells, sensor technologies, and solid-state lighting.

That’s a tall order, but the ASU team says their erbium material is up to it. In fact, erbium is already augmenting these things. Erbium atoms are generally used to dope fiber optic cable, boosting its optical properties and amplifying signals. But because of the particular properties of erbium, cramming enough atoms onto a cable to make it an effective amplifier requires a fairly long cable.
So how do you cram more erbium atoms into a cable? You make the cable itself out of erbium. That’s easier said than done, and the breakthrough here is the erbium compound that can be produced in high quality, single-crystal form. Using the compound, the researchers can create objects with 1,000 times more erbium atoms in them than they could when they were simply doping other materials with erbium. And while that doesn’t translate directly into cables or silicon chips that are 1,000-times faster, it does translate into remarkable improvements in speed and efficiency, the researchers say.


It also enables erbium atoms to be packed into small architectures where they couldn’t be packed in significant numbers previously. That means they can be integrated into silicon chips to speed the performance of computers and other devices even as the fiber optic cables that feed those devices data are also improved by the erbium compound. And all of that could be powered with vastly more efficient PV solar cells made of the erbium compound.

The researchers are testing the material for a range of applications, including those mentioned above. There’s no word yet as to when it might be commercially available. But we imagine it will be fast.

Orbiting Robot Will Gather Space Junk and Turn It Into New Satellites

Our growing space junk problem could become an orbiting spare satellite parts sale if DARPA has its way. The DoD’s research arm has launched a new program, appropriately titled Phoenix, to create new satellites from the decommissioned and dead satellites currently sitting idle in geosynchronous orbit some 22,000 miles above the Earth.

Defunct satellites are in relative abundance in “graveyard orbits” where they have been placed to essentially stay out of the way of newer, working satellites. But while these satellites are out of fuel and out of power, they still retain functioning elements that are both useful and heavy--particularly their antennas. DARPA envisions a scheme where the inner guts of a satellite could be launched into space (at a relatively low cost, since the communications large antennas make up a decent portion of a satellite’s launch weight) and then be attached to an antenna already in orbit.
To do this, DARPA will need an orbiting space robot, or a “tender” spacecraft, that can make all of this happen on orbit. Like an automated mechanic, the tender would rendezvous with “satlets” (those are the small satellite “brains”) that would likely hitch rides on other satellite launches. It would then remove the satlets from the larger satellite or launch vehicle and carry the satlets to a dead satellite where it would attach the satlet to a used antenna and then cut the antenna loose from its former satellite. The tender would then place the new satlet-antenna combo in its new orbit. Just like that, the tender will have built a new satellite from old parts already available in space, limiting the amount of new debris introduced into orbit.


Of course, all of this is going to require some pretty big technological leaps, and as is often the case with DARPA, it’s the derivative technologies that could make the biggest impact here. DARPA is going to need, first and foremost, a space robot with keen machine vision and some degree of AI autonomy, the ability to refuel on orbit (possibly at the International Space Station, which is testing its own satellite refueling tech), and the ability to chase down and intercept satellites as they are orbiting (can’t think of a defense application there).

All said, if DARPA can pull this one off the agency is going to have a nice suite of orbital technologies at its disposal. Not that it would dream of using them for anything other than the originally stated purpose.

DARPA's 'Flying Humvee' Is Moving Ahead, Ready For Prototype

A flying car in mid-2015? There are no guarantees in the world of envelope-pushing, mind-bending military tech, but DARPA says both AAI and Lockheed Martin have produced “feasible designs” for its Transformer (TX) program, known more casually as the “flying Humvee” initiative. Both designs have moved to Phase 2, which requires them to begin work on prototypes for evaluation at the end of fiscal 2012.

This development also brings us a full phase closer to Phase 3, which is where things get really interesting: ground and flight demonstrations, slated for mid-fiscal 2015 should Phase 2 come off as planned. That’s just four years from right now. Given that we’ve been waiting for our cars to take flight for a century now, four years seems pretty reasonable by comparison.
You may recall from our relentless fawning over the Transformer program that the flying Humvee needn’t just be a roadable aircraft. Its design parameters call for a durable on- and off-road capable vehicle that can take small arms fire in stride and make a speedy shift from ground vehicle to vertical-takeoff-and-landing aircraft. It also needs to be easy enough to pilot that any grunt with a driver’s license can also grab the sticks of this new flying machine and be expected to pilot it safely (presumably with the help of autonomous computerized flight controls).


So you’re looking at some design incongruities there. It needs to be light enough to lift off vertically under its own power, yet will require a serious power plant (weight) and some kind of wings and/or propellor/ducted fan (more weight). It also needs space for all this extra hardware while still maintaining passenger space for four soldiers (including driver).

Like most of DARPA’s challenges, bringing this one to life won’t be simple. But speaking to a conference recently, DARPA Transformer program manager Stephen Waller told attendees that “we are seeing feasible designs,” Aviation Week’s Ares blog reports. That’s promising. Let’s just hope these things can actually get off the ground.

A Texas Sheriff's Department is Launching an Unmanned Helodrone that Could Carry Weapons

A sheriff’s office outside of Houston is taking a big and potentially controversial step forward with a new piece of law enforcement technology. The Montgomery County Sheriff’s Office in Conroe, Texas, is prepping its deputies to fly a $300,000 unmanned ShadowHawk helicopter --paid for with a Department of Homeland Security grant--that someday might carry a weapons payload.

This wouldn’t mark the first time a law enforcement agency has put a drone in the air, but the potential for carrying anything besides a surveillance payload is unprecedented as far as we’ve heard. There won’t be any Hellfire missiles here, but ShadowHawk--built locally by a company called Vanguard Defense Industries--is designed to carry a range of less-lethal payloads, including Taser-esque weapons that deliver an electric charge or a firearm that fires beanbag rounds known as stun batons.
Naturally, the very idea of this is fraught with controversies. Some local politicians and citizens are worried about invasions of privacy, while others are concerned that the drone will be put to use catching speed limit violators, an idea that caused some controversy in the Houston area back in 2007 (the idea was later scrapped).


Sheriff Tommy Gage told local media in Houston that his department’s ShadowHawk would only be deployed in emergency circumstances (like a hunt for a missing person) or in situations where sheriff’s deputies know they are dealing with law breakers, such as when police are actively pursuing suspects on the ground or in hostage situations. It could also be deployed to aid area firefighters in wildfire situations (nearby Bastrop was ablaze for several days this summer, and wildfires have become an increasing problem in Texas in recent years).

But weaponized law enforcement drones? Even loaded with a non-lethal payload the Montgomery Country Sheriff’s Department is dipping its toes in uncharted waters. Texas perhaps has a reputation for dealing unsympathetically with lawbreakers, but Texans are also fiercely protective of their civil liberties. There no telling how this one will shake out, but if Montgomery County can make a success of its drone deployment this could mark the beginning of a larger trend in law enforcement and a larger, more hands-on role for robotics.

The Air Force's 'Micro-Aviary' Gives Tiny Flying Robots a Place to Call Home

We have a lot of love for microdrones here at PopSci--everything from bird-like flapping wing drones to cyborg insects controlled by microcomputers--so we’re thrilled to see the Air Force is showing them some love as well. The Air Force Research Lab has build a “Micro-Aviary” at Wright Patterson AFB in Ohio where tiny flying robots will be the central focus. And aside from being drone-centric, it is one sweet sensor-filled laboratory.

The Micro-Aviary will specialize in what the DoD has deemed the next generation of intelligence and military robot capabilities: tiny drones that are largely indistinguishable from insects or birds that can surreptitiously move about undetected, performing surveillance and intelligence gathering missions or even delivering payloads like tracking devices or even weapons. But first, the military has to build them.

At the Micro-Aviary, tiny helicopters will mix with robot dragonflies and hovering robo-hummingbirds in a space walled in by motion sensors that can track the drones that fly there to within about a tenth of an inch. Data from those sensors will not only evaluate the performance of the drones that are there but inform the design of future generations of flapping-wing drones and stealthy robot micro-helos. Tale an inside look at the lab below.

Video: Japanese Robotic Polar Bear Gently Smacks Snorers in the Face

Talk about a rude awakening. If you snore, this new pillowbot from Japan will gently brush your cheek to get you to stop — or flip out of bed in terror as its disturbingly slow arm moves toward you.

It’s designed to help people sleep better by stopping chronic snorers and those who suffer from sleep apnea, which causes breathing difficulty while sleeping. The robot, called “Jukusui-kun” or “deep sleep” in Japanese, is designed to look like a friendly snoozing polar bear. It is connected to a small glove device (also fuzzy bear-shaped) that measures blood oxygen levels, and a below-the-sheets sensor that detects loud noises. The pillow itself also has a microphone to monitor snore decibel levels. A person’s vital stats are pre-programmed into a terminal, which connects wirelessly so you don’t get tangled up in cables.

When the sensors detect blood-oxygen levels are getting too low, or when snoring becomes unbearably loud (yeah that’s right), the bear-pillow’s paw moves slowly and frighteningly toward the sleeping person’s face. This gentle cheek-brush induces the snoring person to turn over on to his or her side, which stops snoring and restores a more restful sleep. The bearpillowbot was developed at Tokyo's Waseda University and was unveiled during the International Robot Show.

Invented: World's Lightest Material, 99.99 Percent Air

A collaboration of researchers from HRL, CalTech, and UC Irvine have created the new world's lightest material--some 100 times lighter than styrofoam. It's even lighter than aerogel, one of our favorite ultralight materials.

The material is a micro-lattice in structure, with the 0.01 percent of the material that's solid consisting of hollow tubes that are only 100 nanometers thick. It's rated at a density of 0.9 mg/cc, lighter than even the lightest aerogels, which have only achieved 1.1 mg/cc. It's also extraordinarily strong and shock-absorbent, thanks to all that air: it can compress by 50 percent and completely recover its shape, highly unusual for a material that is essentially metallic. It was actually inspired by architectural structures rather than other ultralight materials--the team looked to the Golden Gate Bridge and the Eiffel Tower to see how those structures are so light and yet so strong.

The project was undertaken for, who else, DARPA, which says it could be used for products ranging from battery electrodes to energy damping in addition to insulation, the main use for prior lightweight champ aerogel.

Cheaper LEDs

Flexible arrays of bright inorganic LEDs could mean cheaper displays and lighting.
A new technique makes it possible to print flexible arrays of thin inorganic light-emitting diodes for displays and lighting. The new printing process is a hybrid between the methods currently used to make inorganic and organic LEDs, and it brings some of the advantages of each, combining the flexibility, thinness and ease of manufacturing organic polymers with the brightness and long-term stability of inorganic compounds. It could be used to make high-quality flexible displays and less expensive LED lighting systems.

Inorganic LEDs are bright and long lasting, but the expense of manufacturing them has led to them being used mainly in niche applications such as billboard-size displays for sports arenas. What's more, the manufacturing process for making inorganic LED displays is complex, because each LED must be individually cut and placed, says John Rogers, a materials science professor in the Beckman Institute at the University of Illinois at Urbana-Champaign. So display manufacturers have turned to organic materials, which can be printed and are cheaper. While LED-based lighting systems are attractive because of their low energy consumption, they remain expensive. The new printing process, developed by Rogers and described today in the journal Science, could bring down the cost of inorganic LEDs because it would require less material and simpler manufacturing techniques.

Displays based on inorganic LEDs, says Nicholas Colaneri, director of the Flexible Display Center at Arizona State University in Tempe, "are not generally economical to make." The manufacturing process involves sawing wafers of semiconducting materials such as gallium arsenide, picking and placing each piece individually using robotics, and adding electrical connections one at a time.
To make the hybrid LEDs, the Illinois researchers start by growing an inorganic semiconducting material on top of what Rogers calls a "sacrificial" layer. The group uses a chemical bath to etch out LEDs that are just 10 to 100 micrometers on each side. Each LED is then secured with polymer anchors on two of its four corners. The anchors hold the LED in place during a second chemical bath that undercuts the LED, removing the sacrificial layer. The LEDs, which are about 2.5 micrometers thick, can then be picked up on a soft stamp and printed onto a glass, plastic or rubber substrate covered in a polymer adhesive. "You can deliver thousands of LEDs in a single step," says Rogers. "And because they're so thin, they can be interconnected using the conventional processes" used for organic LEDs and liquid-crystal displays.

Smarter LED Lights

A new approach to LED lighting uses network cables, rather than conventional electrical wiring, to supply power to lights. Developed by a startup in Fremont, CA, the system also allows the cables to carry data from an array of sensors on the lights to a central control station. The system would cost about the same as a conventional lighting system, but because it can sense and control every light in a building, it could cut power consumption from lighting by 50 to 80 percent.

The new system offers a better way to control LEDs, which are relatively efficient and long-lasting compared to conventional lights, by taking advantage of the fact that they run on low-voltage direct current power. Current LED-based systems require transformers at each light to convert the higher-voltage alternating current in conventional wiring into lower-voltage direct current. The new system converts alternating current to low-voltage direct current at a central location, rather than at each light. This more efficient method cuts energy consumption by 10 to 20 percent, according to Jeremy Stieglitz, vice president of marketing for Redwood Systems, which will start selling its systems this summer.

The remaining energy savings come from using sensors and a central controller to reduce light use. The company has also developed a method for using those same power cables to carry data. Each LED can be fitted with inexpensive sensors that can be used to optimize light levels and ensure the lights are operating efficiently. Such sensors can also provide detailed information about temperature and where people are in the building--information that can be used to control heating and cooling systems. The sensing and controls, says Steiglitz, add very little cost to the new system because the network connections and power supply for the sensors are already in place.

Each light comes equipped with six sensors. Two are similar to what's used in some newer lighting systems--they detect motion and ambient light (used to turn off lights when there's enough daylight). But where conventional systems control all the lights for an entire room or open cubicle area, the new system allows for control at each light. So the system could, for example, compensate for lower daylight levels further from windows, or dim lights in a large space where no one is working. The new system also monitors task lighting with a third sensor, to ensure that desktops are receiving enough light (something individuals could set according to their preference).
The remaining three sensors can optimize the efficiency of LEDs and help control heating and cooling. A voltage and current sensor detects how much power each LED is using. A temperature sensor on the LED itself, along with one that measures ambient temperatures, tells the central controller if the light is operating at ideal temperatures.

The new system could lead to "huge cost savings," particularly for installing lighting for new buildings, and it could improve the consistency of lighting in a building, says Avraham Mor, a partner in the lighting design company Lightswitch, based in Chicago. But he says installers will have to be educated on the system, and it could be difficult to convince an entrenched lighting industry to switch to new

Lighting Sheets Made of Tiny LEDs

A company called Nth Degree Technologies hopes to replace light bulbs with what look like glowing sheets of paper. The company's first commercial product is a two-by-four-foot-square light, which it plans to start shipping to select customers for evaluation by the end of the year.

The technology could allow for novel lighting designs at costs comparable to the fluorescent light bulbs and fixtures used now, says Neil Shotton, Nth Degree's president and CEO. Light could be emitted over large areas from curved surfaces of unusual shapes. The printing processes used to make the lights also make it easy to vary the color and brightness of the light emitted by a fixture. "It's a new kind of lighting," Shotton says.

Nth Degree makes its light sheets by first carving up a wafer of gallium nitride to produce millions of tiny LEDs—one four-inch wafer yields about eight million of them. The LEDs are then mixed with resin and binders, and a standard screen printer is used to deposit the resulting "ink" over a large surface.

In addition to the LED ink, there's a layer of silver ink for the back electrical contact, a layer of phosphors to change the color of light emitted by the LEDs (from blue to various shades of white), and an insulating layer to prevent short circuits between the front and back. The front electrical contact, which needs to be transparent to let the light out, is made using an ink that contains invisibly small metal wires.
The new transparent electrical contact could itself prove important as a replacement for the indium tin oxide (ITO) used in touch screens and other displays. ITO is brittle and can't be printed, so it's not suitable for flexible displays. It can also be expensive, depending on the price of indium.

While the devices the company has made so far are more efficient than incandescent lights, they're not yet as efficient as fluorescent lights. They emit 20 lumens per watt, compared with about 80 lumens per watt for typical overhead fluorescent lights and 65 lumens per watt for compact fluorescents. A 60-watt light bulb from GE gets about 14 lumens per watt.

Gates-backed Liquid Metal Battery hires CEO

Liquid Metal Battery, a company formed to make cheap storage for wind and solar power, has hired its first CEO.

Phil Giudice, who was the third employee of demand-response company EnerNoc and the Massachusetts Department of Energy Resources Commissioner until earlier the year, announced his "new gig" on Twitter. One of his tasks as CEO is to raise more money to build up the company, he told The Boston Globe.

Liquid Metal Battery was spun out of the lab of Donald Sadoway, a professor of materials chemistry at the Massachusetts Institute of Technology. Funding for the company has come from France-based oil giant Total and software tycoon Bill Gates, who took an interest in the technology after watching Sadoway's lectures online. Sadoway's lab has also received funding from the Department of Energy's ARPA-E research program.

Liquid Metal Battery is taking a radically different approach from lithium ion or other conventional batteries in pursuit of a low-cost system for storing many hours of renewable energy.
The active components in the battery--the anode, the cathode, and electrolyte--are liquid metal alloys, an approach that promises to make the batteries durable for many years. To get to a liquid state, the metals will be held at high temperatures between 400 degrees and 700 degrees Celsius. Its first prototype system is about the size of four pizza boxes, is using abundant salts, and is expected to use metals such as antimony and magnesium. Rather than string thousands of small battery cells together, the company's storage systems will be much larger and able to reduce the number of cells by 50 to 100 times, according to Sadoway.

Using manufacturing techniques similar to aluminum smelting, the company's goal is to make the batteries cheap enough to be used with renewable-energy systems. Executives have said the goal is to get in the range of $250 per kilowatt-hour, which is substantially less than half the price of today's lithium ion batteries.

Auto-dimming LED tech gives new meaning to daylight savings

Digital Lumens is using daylight to ratchet up the efficiency of its commercial LED light fixtures.

The company today is announcing a new product line that boosts the amount of light each fixture can give off and improves energy efficiency by up to 50 percent with an embedded daylight sensor.

Efficient LED lighting promises to bring significant energy savings to commercial and industrial customers because businesses tend to spend more money on lighting. But just like the consumer LED lightbulb market, the cost of LED fixtures is still higher than other lighting technologies.

Digital Lumens' fixture, which is about the size of a desktop PC, combines an array of LED light sources with a networking chip. This connects fixtures to a centralized management application to schedule and control lights, thus improving efficiency. The product is designed for warehouse aisles and manufacturing spaces, and it pays back the initial higher cost within a year or two, according to the company.

The fixture's daylight sensor enables it to gradually dim itself when there's daylight available, leading to energy savings between 25 percent and 50 percent, according to the company.

Engineers decided to use digital sensors normally used for automatically dimming screens on cell phones, said Joseph Adiletta, senior product manager. Photo light sensors are normally added to an area after fixtures are put up, rather than being integrated into each fixture, he said.

Digital Lumens' latest products also give off brighter light, between 10,000 and 26,000 lumens. That makes them suitable for a broader set of uses--and competitive with high-intensity discharge or fluorescent lamps--but by producing between 81 and 85 lumens per watt, they are more energy-efficient, said Michael Feinstein, vice president of sales and marketing.

Robot Venus flytraps could eat bugs for fuel

ROBOTS that mimic the Venus flytrap could run on live insects and spiders, snatching and digesting them for fuel. Now two prototypes have been developed that employ smart materials to rapidly ensnare their prey.

Venus flytraps (Dionaea muscipula) catch insects using two specially adapted leaves. When a bug lands it brushes tiny hairs on the surface, triggering the trapping mechanism. The leaves snap shut in a mere 100 milliseconds, and the plant kills and digests its quarry (see diagram).

Recreating this method means finding materials that can not only detect the presence of an insect but also close on it quickly. At Seoul National University in South Korea, Seung-Won Kim and colleagues have done this using shape memory materials. These switch between two stable shapes when subjected to force, heat or an electric current.

The team used two different materials - a clamshell-shaped piece of carbon fibre that acts as the leaves, connected by a shape-memory metal spring. The weight of an insect on the spring makes it contract sharply, pulling the leaves together and enveloping the prey. Opening the trap once more is just a matter of applying a current to the spring.

Mohsen Shahinpoor at the University of Maine in Orono took a different approach. His robot flytrap uses artificial muscles made of polymer membranes coated with gold electrodes. A current travelling through the membrane makes it bend in one direction - and when the polarity is reversed it moves the other way.

Bending the material also produces a voltage, which Shahinpoor has utilised to create sensors. When a bug lands, the tiny voltage it generates triggers a larger power source to apply opposite charges to the leaves, making them attract one another and closing the trap (Bioinspiration and Biomimetics, DOI: 10.1088/1748-3182/6/4/046004).

"We should be able to benefit enormously from these flytrap technologies," says Ioannis Ieropoulos of the Bristol Robotics Lab in the UK. He and colleagues previously developed Ecobot, a robot that can digest insects, food scraps and sewage to power itself. Ecobot uses bacteria to break down a fly's exoskeleton in a reaction that liberates electrons into a circuit, generating electricity.

But without a way to catch prey, the researchers either manually feed Ecobot with dead flies or use an ultraviolet bug lure - like those used in restaurants. That's no good for an autonomous robot, though. What's more, UV lures need to be on all the time, wasting precious power, says Ieropoulos. "We'd be happy to talk to these groups about their flytraps."

Surveillance robots know when to hide

The creation of robots that can hide from humans while spying on them brings autonomous spy machines one step closer

THE spy approaches the target building under cover of darkness, taking a zigzag path to avoid well-lit areas and sentries. He selects a handy vantage point next to a dumpster, taking cover behind it when he hears the footsteps of an unseen guard. Once the coast is clear, he is on the move again - trundling along on four small wheels.

This is no human spy but a machine, a prototype in the emerging field of covert robotics. It was being put through its paces at a demonstration late last year by Lockheed Martin's Advanced Technology Laboratories at Cherry Hill, New Jersey. With an aerial drone to their credit (see "Unseen watcher in the sky"), the company now wants to design autonomous robots that can operate around humans without being detected.

What makes the robot special is its ability to build a computer model of its surroundings, incorporating information on lines of sight. The robot is fitted with a laser scanner to allow it to covertly map its environment in 3D. It also has a set of acoustic sensors which it uses to distinguish nearby footsteps and their direction.

Lead engineer Brian Satterfield says the robot was designed to operate within four constraints: "Avoiding visible detection by sentries of known locations, avoiding potential detection by sentries whose positions were unknown, avoiding areas in which the robot would have no means of escape, and, as this robot was designed to run at night, avoiding areas that were well lit." To make it hard to spot in the dark, the robot was painted black.

If the robot believes it is in danger of being detected by an approaching sentry, it will try to get to a place where it can hide, Satterfield says. His comment is an example of how natural it is for us to talk about such robots as if they understand how they are perceived and have a "theory of mind".

"Lockheed Martin's approach does include a sort of basic theory of mind, in the sense that the robot makes assumptions about how to act covertly in the presence of humans," says Alan Wagner of the Georgia Institute of Technology in Atlanta, who works on artificial intelligence and robot deception.

But the level at which the robot's software operates is probably limited to task-specific instructions such as, "if you hear a noise, scurry to the nearest dark corner", he says. That's not sophisticated enough to hide from humans in varied environments.

"Significant AI will be needed to develop a robot which can act covertly in a general setting," Wagner says. "The robot will need to consider its own shape and size, to have the ability to navigate potential paths, [to be aware of] each person's individual line of view, the impact that its movement will have on the environment, and so on."

Satterfield's robot was built with off-the-shelf components. Both he and Wagner say that specialised hardware which is more compact and quieter will improve future robots' mobility and their ability to stay hidden. "There are very few fundamental limits that would prevent robots from eventually conducting extended covert missions and evading detection by humans," Satterfield says.

Lockheed Martin's work looks ready to emerge, albeit quietly, into the real world. The US army recently solicited proposals for a "persistent surveillance" robot with concealment capabilities and suited for extended deployments. Later this year, the US Department of Defense is expected to back that up with cash awards for working designs

Lasers Used to Detect Bombs

ScienceDaily (Sep. 17, 2011) — A research team at Michigan State University has developed a laser that could detect roadside bombs -- the deadliest enemy weapon encountered in Iraq and Afghanistan.

The laser, which has comparable output to a simple presentation pointer, potentially has the sensitivity and selectivity to canvas large areas and detect improvised explosive devices -- weapons that account for around 60 percent of coalition soldiers' deaths. Marcos Dantus, chemistry professor and founder of BioPhotonic Solutions, led the team and has published the results in the current issue of Applied Physics Letters.

The detection of IEDs in the field is extremely important and challenging because the environment introduces a large number of chemical compounds that mask the select few molecules that one is trying to detect, Dantus said.

"Having molecular structure sensitivity is critical for identifying explosives and avoiding unnecessary evacuation of buildings and closing roads due to false alarms," he said.

Since IEDs can be found in populated areas, the methods to detect these weapons must be nondestructive. They also must be able to distinguish explosives from vast arrays of similar compounds that can be found in urban environments. Dantus' latest laser can make these distinctions even for quantities as small as a fraction of a billionth of a gram.

The laser beam combines short pulses that kick the molecules and make them vibrate, as well as long pulses that are used to "listen" and identify the different "chords." The chords include different vibrational frequencies that uniquely identify every molecule, much like a fingerprint. The high-sensitivity laser can work in tandem with cameras and allows users to scan questionable areas from a safe distance.

"The laser and the method we've developed were originally intended for microscopes, but we were able to adapt and broaden its use to demonstrate its effectiveness for standoff detection of explosives," said Dantus, who hopes to net additional funding to take this laser from the lab and into the field.

This research is funded in part by the Department of Homeland Security. BioPhotonic Solutions is a high-tech company Dantus launched in 2003 to commercialize technology invented in a spinoff from his research group at MSU.